Function synthesizeWithOpenAi

apps/sim/app/api/tools/tts/unified/route.ts:327–365

(
  params: OpenAiTtsParams
)

})

async function synthesizeWithOpenAi(
  params: OpenAiTtsParams
): Promise<{ audioBuffer: Buffer; format: string; mimeType: string }> {
  const { text, apiKey, model = 'tts-1', responseFormat = 'mp3', speed = 1.0 } = params
  const voice = (params.voice || 'alloy') as OpenAiTtsParams['voice']

  const response = await fetch('https://api.openai.com/v1/audio/speech', {
    method: 'POST',
    headers: {
      Authorization: `Bearer ${apiKey}`,
      'Content-Type': 'application/json',
    },
    body: JSON.stringify({
      model,
      voice,
      input: text,
      response_format: responseFormat,
      speed: Math.max(0.25, Math.min(4.0, speed)),
    }),
  })

  if (!response.ok) {
    const error = await readTtsErrorJson(response, 'OpenAI TTS error response')
    const errorMessage = getTtsErrorMessage(error, response.statusText)
    throw new Error(`OpenAI TTS API error: ${errorMessage}`)
  }

  const audioBuffer = await readResponseToBufferWithLimit(response, {
    maxBytes: MAX_TTS_AUDIO_BYTES,
    label: 'OpenAI TTS audio response',
  })
  const mimeType = getMimeType(responseFormat)

  return {
    audioBuffer,
    format: responseFormat,
    mimeType,
  }
}

async function synthesizeWithDeepgram(
  params: DeepgramTtsParams